1. Field of the Invention
The present invention relates generally to the field of seismic exploration and, more particularly, to methods of processing 2-D and 3-D seismic data to improve the ability of this data to correctly image the structure and stratigraphy of the subsurface geology. Most particularly, it relates to the use of time slices or depth slices to select velocity fields for imaging multi-directional fractures in a homogeneous basement.
2. Description of the Prior Art
In the field of seismic exploration, it is generally well known that hydrocarbon deposits are most commonly found in sedimentary rocks. More dense types of basement, such as granite rocks were commonly avoided because they were not thought to contain hydrocarbon accumulations within the rock mass itself. More recently, it has been discovered that hydrocarbons may reside in fractures of such basement rock. It is a difficult problem in seismic exploration to image these fractures using the methods of the prior art because they are often a small scale feature and the signal to noise ratio in the basement rock is not at a level where these fractures are easy to identify. For example, if the granitic basement rock is highly fractured, there can be much interference to the seismic response of an individual fracture. This will, in many instances, be caused by the migration velocity field being either too fast or too slow and thus not properly collapsing the hyperbolic seismic response generated by the nearby faults and fractures. The exploration problem leading to the development of the procedure described in this patent application was how to obtain a better image of a fractured or faulted lithological unit, especially in 3-D sections of hard basement rock.
As illustrated in FIG. 1, in the oil and gas industry, seismic prospecting techniques are commonly used to aid in the search for and evaluation of subterranean hydrocarbon deposits. Typically, the goal of seismic prospecting is to construct a 2-D or 3-D representation of a formation in order to identify features that are indicative of hydrocarbons. Seismic prospecting generally consists of three separate stages, namely data acquisition, data processing, and data interpretation. The success of a seismic prospecting operation depends on satisfactory completion of all three stages.
In the first state of seismic prospecting, a seismic source (S) is used to generate a seismic signal that propagates into the earth and is at least partially reflected by subsurface seismic reflectors (i.e., interfaces between underground formations having different elastic properties). These reflectors most often arise as the boundaries between two different types of subsurface strata, and as such, are fairly easy to recognize with the current seismic prospecting techniques of the prior art. In any event, the reflected signals (known as “seismic reflections”) are detected and recorded by seismic receivers (R) located at or near the surface of the earth, thereby representing a seismic survey of the subsurface. In a two-dimensional (2-D) seismic survey as shown in FIG. 1, the recording locations are generally laid out along a single straight line, whereas in a three-dimensional (3-D) survey (see FIG. 3) the recording locations are distributed across the surface in a grid pattern. In simplest terms, a 2-D seismic line can be thought of as giving a cross sectional picture, called a “vertical slice” or “vertical section” of the earth layers as they exist below the approximate center point between the shot (S) and geophone (R) location. A 3-D survey, on the other hand, produces a data “cube” or volume that is, at least conceptually, a 3-D representation of the subsurface that lies beneath the survey area. In any event, the recorded signals, or seismic energy data, can then be processed to yield information relating to the subsurface formations identifying such structures as subsurface formation boundaries.
The seismic energy recorded by each seismic receiver during the data acquisition stage is known as a “trace”. A modem seismic trace is a digital recording (analog recordings were used in the past) of the acoustic energy reflecting back from inhomegeneties, or reflectors, in the subsurface, a partial reflection occurring each time there is a change in the acoustic impedance of the subsurface materials. The digital samples are usually acquired at 0.002 second (2 millisecond or “ms”) intervals, although one millisecond and four millisecond sampling intervals are also common. Thus, each digital sample in a seismic trace is associated with a travel time, and in the case of reflected energy, a two-way travel time from the surface to the reflector and back to the surface again. Further, the surface location of every trace in a seismic survey is carefully recorded and is generally made a part of the trace itself (as part of the trace header information). This allows the seismic information contained within the traces to be later correlated with specific subsurface locations, thereby providing a means for posting and contouring seismic data—and attributes extracted therefrom—on a 2-D or 3-D map (i.e., “mapping”).
Once the seismic data has been acquired, it is then processed during the second stage of seismic prospecting so that the traces will accurately represent the subsurface. This processing typically involves the use of various mathematical algorithms that are applied to the data to enhance its signal content and generally make it more amenable to interpretation. One of the main objectives of the data processing stage is to remove or at least attenuate various unwanted seismic energy that accompanies the primary seismic signal when it is recorded. This unwanted energy is typically referred to as “noise”, and represents such things as spurious energy traveling through the formation other than primary signal. Various sources of noise include surface noise, i.e., energy arising from surface equipment, instrumentation, weather and the like, as well as sub-surface noise such as multiples, ghosts, refracted signals and the like. Through processing, the strength of the primary signal energy represented in a trace can be enhanced, while the strength of the unwanted noise energy can be weakened, thus increasing the signal-to-noise ratio, or S/N. As mentioned above, noise is particularly significant when mapping fractures within hard, homogeneous basement rock such as granite because the fractures will often not generate a high enough signal to be recognized without some type of signal to noise enhancement. Also there will be a lack of coherent reflectors relating to lithology in such a rock mass because the rock mass does not have the strata that are prevalent in a sedimentary rock mass.
One method for attenuating unwanted noise components in seismic data traces is through the common-midpoint (CMP) “stacking” process as illustrated in FIG. 2. As is well known to persons skilled in the art, the “midpoint” for a seismic data trace is a point generally close to midway between the source location and the receiver location for that trace (see FIG. 1). According to the CMP method, the recorded seismic data traces are sorted into common-midpoint gathers, as shown in “Record 1” of FIG. 2, each of which contains a number of different seismic data traces having the same midpoint but different source-to-receiver offset distances. Prior to stacking, the seismic data traces within each CMP gather must be corrected for various effects such as normal moveout, discussed below. An example of corrected traces is shown in “Record 2” of FIG. 2. These corrected CMP gathers are then summed or “stacked” together yielding a single stacked data trace which is a composite of the individual seismic data traces in the CMP gather. Through stacking of CMP traces, the unwanted energy on any one trace, see “Record 3” of FIG. 2, is reduced when the trace is added with other CMP traces, such that the desired energy is enhanced. Typically, the stacked data trace has a significantly improved signal-to-noise ratio compared to that of the unstacked seismic data traces.
Normal moveout (NMO) is the increase in reflection time of a trace due to an increase in the offset distance between the source and the receiver detecting the signal. As described above, in a CMP gather, each trace generated by a source-receiver pair has an offset that is the distance from the source to the receiver, an azimuth that is the compass direction between the source and receiver, and a midpoint halfway between the source and receiver. As the traces for a CMP gather are plotted adjacent to one another, the reflector along the various traces appears at later points on the time scale because of the longer offsets as the shot and receiver move apart, resulting in a downward curving hyperbolic pattern characteristic on the plot. The time-domain effect of varying offset must be taken into account in the CMP stack process in order for the signal portions of traces of varying offset to properly align and provide an accurate indication of the depth of the reflector. This is typically handled by way of “normal moveout correction” in which the traces corresponding to source-receiver paths of various lengths are time-shifted relative to one another so that their detected reflection events are aligned in time. The amount of the time shift for a given trace will, of course, depend upon its offset distance. It will also depend upon the velocity with which the acoustic energy travels in the strata along the shot-receiver path. Normal moveout correction therefore requires the estimation or determination of a velocity, commonly referred to as the “stacking” velocity, for deriving the necessary time shift as a function of offset. The relationship between offset and the NMO time correction for a given seismic reflection event in a seismic trace, follows the well-known NMO equation:TX2=To2+(X2/VS2)where To is the zero offset reflection time of the reflection, X is the offset distance of the trace being corrected, VS is the stacking velocity for the reflection event, and TX is the time at offset X for the reflection event in the trace being corrected.
Conventional derivation of the stacking velocity VS for NMO correction is typically done in a “best-fit” manner on the hyperbolic curve described above in order to optimize the accuracy of the correction among all of the traces in the gather. For example, a series of corrections based upon multiple trial stacking velocities may be applied to a gather of traces.
Another phenomenon that must be considered in processing data for production of seismic sections useful for analysis is the effect of dip in the subsurface structures. If the subsurface reflecting structure is flat and horizontal, the reflecting point is directly under the midpoint of the source-receiver pair. This is true for all traces in a CMP gather for that point. However, if the reflecting surface exhibits a dip, the reflecting point for a single trace will generally be shifted along the reflecting surface, and will not appear directly under the source-receiver midpoint. The process which alleviates this problem is called migration.
As is well known to those skilled in the art, the dip and location of a reflector on an unmigrated seismic section is rarely representative of the true dip and subsurface location of the structural or stratigraphic feature that gave rise to that reflector. Except in the case where the subsurface consists of homogeneous nearly-horizontal layers, the recorded seismic expression of a structural or stratigraphic event must be migrated before it can be reliably used to locate subsurface features of interest. In areas of steep dip, a reflection that is apparently located directly below a particular shot point may actually be found several hundreds of feet laterally away from that shot point. Additionally, in complex structural areas where faulting, severe asymmetrical folding and sharp synclines are present, diffractions and multiple reflections may interfere with reflections from the primary reflectors to the point where the resulting seismic section bears little or no resemblance to the actual subsurface structure.
Broadly speaking, migration improves a seismic section or volume by “focusing” the seismic data contained therein. Migration improves the seismic image by correcting the lateral mispositioning of dipping seismic reflectors; collapsing diffractions caused by point scattering centers and subsurface fault terminations; resolving crossing reflectors (conflicting dips); and improving the vertical and lateral resolution of the seismic data, among many others. As a general rule, seismic data that have been properly migrated reveal an enhanced or a truer picture of the subsurface than unmigrated seismic data, the ultimate goal of seismic migration being to produce a seismic section or volume that accurately represents the configuration of the geology of the subsurface.
There are two broad variants of seismic migration: migration after stack (post-stack) and migration before stack (prestack). Post-stack migration is applied, as the name suggests, to seismic traces after they have been stacked—a stacked seismic trace being one that is formed by combining together two or more traces to form a single composite trace. Prestack migration, on the other hand, is applied to seismic traces before they have been stacked. Other things being equal prestack migration is always preferred—both theoretically and in practice—because it has the potential to produce a more accurate picture of the subsurface stratigraphy and structure. However, the computational effort involved in computing a prestack migration is many times that required to do a post-stack migration. For 2-D seismic lines, this additional computational effort is generally manageable and prestack migration is often applied to seismic data that has been taken in areas where the subsurface structure is complicated. Also the increased computer power now available has enabled 3-D prestack time and depth migrations to be used on a fairly regular basis.
A further division of seismic migration algorithms may be made based on whether the migration takes place in “depth” or “time.” Broadly speaking, the operational difference between the two approaches is that a trace in the seismic volume that results from a depth migration contains digital samples that are separated by units of depth, rather than time. Whereas a seismic trace in a conventional time seismic survey has samples that are spaced some distance Δt apart in time (where, for example, Δt might be 4 ms), successive samples in a depth migrated volume are spaced some depth, say Δz, units apart. Prestack and post-stack migrations may be performed either in time or in depth. Finally, it is well known to those skilled in the art that seismic data that have been migrated in depth may be transformed into a conventional “time” section and vice versa.
In any event, whether the seismic data processing involves NMO correction, migration or some other type of corrective function, the first step in the corrective process is to identify a velocity field that is representative of the signal velocities as the signal travels through the subsurface geology. “Velocity picking” or “velocity analysis” in seismic exploration refers to the selection of one or more velocity functions, usually time or depth dependent, or velocity volumes in order to perform these various mathematical functions on the seismic data.
Generally velocity analysis involves making decisions as to multiples, overpressured zones, etc. Those skilled in the art will understand that such decisions often do not result in optimal clarity of the final result generally due to unknown factors of the earth's subterranean features. For instance, the step of migration may involve making assumptions about velocity modeling techniques where the actual velocity in the relevant portion of the earth's volume may not be precisely known due to various or complex subterranean features. No matter what velocity determination process is used, however, velocity analysis is based in large part on personal interpretation, and therefore, subject to imaging errors, especially with respect to those formation features that are limited in extent, such as basement fractures.
With respect to velocity analysis for NMO, Equation 1 is often utilized since, for a horizontally layered earth having a single horizontal reflector, seismic signal traveltime bears a hyperbolic relation to source-to-receiver offset distance. Commonly, the velocity of the seismic signal VS is referred to as the “normal-moveout velocity” or “NMO velocity”. However, as is well known in the art, the NMO velocity for a CMP gather is not a constant. Typically, the NMO velocity increases as two-way, zero-offset traveltime increases. For this reason, proper stacking of the seismic data traces within a CMP gather requires knowledge of the NMO velocity as a function of two-way, zero-offset traveltime.
Typically, determination of the NMO velocity function is done manually by expert seismic analysts. One conventional method for determining the NMO velocity function is through the use of velocity spectra. According to this method, the individual seismic data traces in a CMP gather are repeatedly NMO-corrected and stacked using a range of trial velocity values. The resulting amplitude or stack power of the stacked traces are then displayed side-by-side on a plane of velocity versus two-way, zero-offset traveltime, producing what is known as the “velocity spectrum”. The velocity which results in the highest stacked amplitude for a given reflection is selected or “picked” as the NMO velocity for that reflection. The NMO velocity function may then be expressed as a set of velocity-traveltime pairs. Thus, while many other conventional methods for determining or “picking” the NMO velocity function are known, the goal is to select a velocity model or construct a velocity field that corresponds to the best lineup of the signal along the hyperbolic trajectory over the entire CMP gather. In any event, it should be apparent that the velocity analysis relating to NMO is crucial to proper NMO time correction.
Likewise, it is necessary to define a velocity model before a seismic migration can be performed. Again, the velocity model is just a specification of the subsurface velocity structure as a function of depth for subsurface points located in the vicinity of the survey. The velocity model might be simple (involving only a few layers) or complex, but it has always been the conventional wisdom that the best velocity model is the one that most accurately represents the actual configuration of the rock units in the subsurface. In fact, the modern trend has been to move toward increasingly accurate velocity models in an effort to improve the final migrated product.
The most popular methods of migration in use today are all based ultimately on a solution to the wave equation. As is well known to those skilled in the art, the wave equation is a second order differential equation that describes seismic wave propagation in the subsurface. To the extent that this equation accurately represents seismic wave propagation in the subsurface—and to the extent that the true subsurface velocity and other rock parameters are correctly specified—this equation can be used to migrate seismic data with considerable accuracy, a solution to the wave equation producing the theoretically correct way to migrate seismic data. Of course, for most media the wave equation cannot be solved efficiently, thus some degree of approximation must be introduced into the solution. As a general rule, the better/more precise the approximation is to the true solution, the better the resulting migration. As might be expected though, the more accurate the approximation, the greater the amount of computer time required to perform the migration. Thus, the most desirable migrations are those that use the highest fidelity wave transmission model, the disadvantage of this sort of models being that the computer time necessary to calculate a migration is lengthened accordingly.
The most common approaches to wave equation (based) migration are finite-difference methods, frequency domain (f-k) approaches, and Kirchhoff integral migration. Each of these methods has its advantages and disadvantages. No matter which migration method is used, however, seismic migration is a computationally intensive process, even for simple velocity models. In the case of 2-D data, modem computer speeds (coupled in some cases with the use of parallel and massively parallel processors) have made post-stack migration a routine processing step for seismic lines, and prestack migration, although correspondingly more costly from a computation standpoint, is also done rather routinely. However, the computational costs associated with a full prestack migration of 3-D seismic data are comparatively high. By way of example, a 3-D survey might consist of one million or more CMPs (e.g., a grid of 1024 CMPs by 1024 CMPs), each of which might be 40 to 60 fold. Thus, as many as 40 to 60 million unstacked seismic traces might be collected in a typical 3-D survey.
Generally, the prior art methods of performing velocity analysis are predominantly performed using vertical slice analysis to determine the velocity field that best flattens a reflector in a common midpoint (CMP) sort or to determine the velocity that gives the strongest stack or migrated response by the use of constant velocity stacks or constant velocity migrations. Typically, such methods use data sorted in the inline and crossline directions to determine these velocities. An example of vertical slices in the inline and crossline directions is illustrated in FIG. 3. The vertical slice technique of the prior art permits the analyst to reasonably determine the velocity at essentially a point on the 2-D line or within the 3-D volume. This type of analysis is somewhat limited as each slice will only image a limited area for fault detection. Heretofore, use of horizontal slice techniques in the prior art have been limited only to use in interpretation. More specifically, horizontal time or depth slice techniques as used in the prior art for 3-D data is only used in the interpretive function. Such data permits the geophysicist to obtain a plan view showing such things as the fault pattern as it exists across an area in the xy plane. This knowledge of the fault pattern in a plan view can aid the geophysicist in determining such things as stress orientation, possible hydrocarbon trapping, etc. However, the drawback to the use of horizontal slices only for interpretation is that velocity changes in the plan view are not readily apparent to the analyst. It has been discovered however, as part of the inventive process described herein, that horizontal time or depth slices can be used in the velocity determination across a 3-D volume, thereby permitting the analyst to more easily identify velocity changes across the 3-D volume.
Thus, there is a need for a method to better image fractures in hard basement rock. Additionally, the method should provide for a highly accurate migration at minimal additional computational cost. Accordingly, it should now be recognized, as was recognized by the present inventors, that there exists, and has existed for some time, a very real need for a method of seismic data processing that would address and solve the above-described problems.